Phi-4 hardware requirements

Open license · MITChatReasoning

About the model

Parameters
14.7B
Attention
Full attention
Max context
16,384 tokens
Released
2024-12-12

GGUF files

Tracked quantizations of Phi-4, smallest first
QuantFile sizeQualityPublished by
Q4_K_Mbaseline9.05 GBGood balanceCommunity GGUF · lmstudio-community
Q6_K12.03 GBNear-losslessCommunity GGUF · lmstudio-community
Q8_015.58 GBNear-losslessCommunity GGUF · lmstudio-community

Memory needed by quant and context

Weights + f16 KV cache + compute buffer, in GB. Add about 0.6 GB if the GPU also drives your display on Windows.
Quant4k8k16k
Q4_K_M10.411.313.1
Q6_K13.414.316.0
Q8_017.017.819.6

Which GPUs can run Phi-4?

How to read the verdicts

Runs great
Fully on the GPU at 20 tok/s or more
Runs well
Fully on the GPU at 8–20 tok/s; MoE experts in system RAM at 20 tok/s or more; or at least 90% on the GPU at 8 tok/s or more
Runs slowly
2–8 tok/s; CPU-only; less than 90% on the GPU; or MoE experts in system RAM below 20 tok/s
Won't run
Does not fit, or under 2 tok/s

Dense models split between GPU and CPU slow down sharply — the CPU side sets the pace. MoE models that keep only their experts in system RAM degrade far more gently.

Scroll sideways to see every column.

54 GPUs, Macs and CPU setups at Q4_K_M with 8k context, sorted by verdict and speed
HardwareVerdictSpeedQuantMemoryRuns asContextNotesPrice
GeForce RTX 5090
Runs great
est. 116.9 tok/s102.9–131.0calibrated estimate ±12%
Q4_K_M11.9 / 32.0 GBFull GPUup to 16k—$4,200 street (as of 2026-08-09)
Radeon RX 7900 XTX
Runs great
est. 69.8 tok/s61.4–78.2calibrated estimate ±12%
Q4_K_M11.9 / 24.0 GBFull GPUup to 16k—$999 launch MSRP
GeForce RTX 3090 Ti
Runs great
est. 65.8 tok/s57.9–73.7calibrated estimate ±12%
Q4_K_M11.9 / 24.0 GBFull GPUup to 16k—$1,999 launch MSRP
GeForce RTX 4090
Runs great
est. 65.8 tok/s57.9–73.7calibrated estimate ±12%
Q4_K_M11.9 / 24.0 GBFull GPUup to 16k—$2,755 street (as of 2026-08-09)
GeForce RTX 5080
Runs great
est. 62.6 tok/s55.1–70.2calibrated estimate ±12%
Q4_K_M11.9 / 16.0 GBFull GPUup to 16k—$1,256 street (as of 2026-08-09)
GeForce RTX 3090
Runs great
est. 61.1 tok/s53.7–68.4calibrated estimate ±12%
Q4_K_M11.9 / 24.0 GBFull GPUup to 16k—$1,050 street (as of 2026-08-09)
GeForce RTX 3080 12GB
Runs great
est. 59.5 tok/s52.4–66.7calibrated estimate ±12%
Q4_K_M11.9 / 12.0 GBFull GPUup to 8k—$799 launch MSRP
GeForce RTX 5070 Ti
Runs great
est. 58.5 tok/s51.4–65.5calibrated estimate ±12%
Q4_K_M11.9 / 16.0 GBFull GPUup to 16k—$949 street (as of 2026-08-09)
GeForce RTX 5080 Laptop
Runs great
est. 58.5 tok/s51.4–65.5calibrated estimate ±12%
Q4_K_M11.9 / 16.0 GBFull GPUup to 16k——
GeForce RTX 5090 Laptop
Runs great
est. 58.5 tok/s51.4–65.5calibrated estimate ±12%
Q4_K_M11.9 / 24.0 GBFull GPUup to 16k——
Radeon RX 7900 XT
Runs great
est. 58.2 tok/s51.2–65.1calibrated estimate ±12%
Q4_K_M11.9 / 20.0 GBFull GPUup to 16k—$899 launch MSRP
GeForce RTX 4080 Super
Runs great
est. 48.0 tok/s42.3–53.8calibrated estimate ±12%
Q4_K_M11.9 / 16.0 GBFull GPUup to 16k—$999 launch MSRP
Radeon RX 9070
Runs great
est. 46.9 tok/s41.3–52.5calibrated estimate ±12%
Q4_K_M11.9 / 16.0 GBFull GPUup to 16k—$639 street (as of 2026-08-09)
Radeon RX 9070 XT
Runs great
est. 46.9 tok/s41.3–52.5calibrated estimate ±12%
Q4_K_M11.9 / 16.0 GBFull GPUup to 16k—$689 street (as of 2026-08-09)
GeForce RTX 4080
Runs great
est. 46.8 tok/s41.2–52.4calibrated estimate ±12%
Q4_K_M11.9 / 16.0 GBFull GPUup to 16k—$1,199 launch MSRP
Radeon RX 7800 XT
Runs great
est. 45.4 tok/s39.9–50.8calibrated estimate ±12%
Q4_K_M11.9 / 16.0 GBFull GPUup to 16k—$499 launch MSRP
GeForce RTX 4070 Ti Super
Runs great
est. 43.8 tok/s38.6–49.1calibrated estimate ±12%
Q4_K_M11.9 / 16.0 GBFull GPUup to 16k—$799 launch MSRP
GeForce RTX 5070
Runs great
est. 43.8 tok/s38.6–49.1calibrated estimate ±12%
Q4_K_M11.9 / 12.0 GBFull GPUup to 8k—$629 street (as of 2026-08-09)
GeForce RTX 5070 Ti Laptop
Runs great
est. 43.8 tok/s38.6–49.1calibrated estimate ±12%
Q4_K_M11.9 / 12.0 GBFull GPUup to 8k——
GeForce RTX 4090 Laptop
Runs great
est. 37.6 tok/s33.1–42.1calibrated estimate ±12%
Q4_K_M11.9 / 16.0 GBFull GPUup to 16k——
Apple M5 Max (40-core GPU) · 48 GB
Runs great
est. 34.3 tok/s27.5–41.2calibrated estimate ±20%
Q4_K_M11.3 / 33.6 GBUnified memoryup to 8k——
Apple M3 Ultra · 96 GB
Runs great
est. 33.6 tok/s26.9–40.3calibrated estimate ±20%
Q4_K_M11.3 / 67.2 GBUnified memoryup to 8k——
GeForce RTX 4070
Runs great
est. 32.9 tok/s28.9–36.8calibrated estimate ±12%
Q4_K_M11.9 / 12.0 GBFull GPUup to 8k—$599 launch MSRP
GeForce RTX 4070 Super
Runs great
est. 32.9 tok/s28.9–36.8calibrated estimate ±12%
Q4_K_M11.9 / 12.0 GBFull GPUup to 8k—$599 launch MSRP
Apple M2 Ultra · 64 GB
Runs great
est. 32.8 tok/s26.2–39.4calibrated estimate ±20%
Q4_K_M11.3 / 44.8 GBUnified memoryup to 8k——
Apple M4 Max (40-core GPU) · 48 GB
Runs great
est. 30.5 tok/s24.4–36.6calibrated estimate ±20%
Q4_K_M11.3 / 33.6 GBUnified memoryup to 8k——
GeForce RTX 5060 Ti 16GB
Runs well
est. 29.2 tok/s25.7–32.7calibrated estimate ±12%
Q4_K_M11.9 / 16.0 GBFull GPUup to 16k—$569 street (as of 2026-08-09)
GeForce RTX 4080 Laptop
Runs well
est. 28.2 tok/s24.8–31.6calibrated estimate ±12%
Q4_K_M11.9 / 12.0 GBFull GPUup to 8k——
Apple M5 Max (32-core GPU) · 36 GB
Runs well
est. 25.7 tok/s20.6–30.9calibrated estimate ±20%
Q4_K_M11.3 / 25.2 GBUnified memoryup to 16k——
Intel Arc B580
Runs well
est. 25.1 tok/s20.1–30.1calibrated estimate ±20%
Q4_K_M11.9 / 12.0 GBFull GPUup to 8k—$290 street (as of 2026-08-09)
GeForce RTX 3060 12GB
Runs well
est. 23.5 tok/s20.7–26.3calibrated estimate ±12%
Q4_K_M11.9 / 12.0 GBFull GPUup to 8k—$250 street (as of 2026-08-09)
Radeon RX 9060 XT 16GB
Runs well
est. 23.4 tok/s20.6–26.2calibrated estimate ±12%
Q4_K_M11.9 / 16.0 GBFull GPUup to 16k—$449 street (as of 2026-08-09)
Apple M4 Max (32-core GPU) · 36 GB
Runs well
est. 22.9 tok/s18.3–27.5calibrated estimate ±20%
Q4_K_M11.3 / 25.2 GBUnified memoryup to 16k——
GeForce RTX 2080 Ti
Runs well
est. 21.6 tok/s17.3–25.9calibrated estimate ±20%
Q4_K_M11.0 / 11.0 GB + 0.9 GB RAMPartial offloadup to 8k—$999 launch MSRP
GeForce GTX 1080 Ti
Runs well
est. 19.1 tok/s13.4–24.8theoretical estimate ±30%
Q4_K_M11.0 / 11.0 GB + 0.9 GB RAMPartial offloadup to 8k—$699 launch MSRP
GeForce RTX 4060 Ti 16GB
Runs well
est. 18.8 tok/s16.5–21.0calibrated estimate ±12%
Q4_K_M11.9 / 16.0 GBFull GPUup to 16k—$499 launch MSRP
Apple M5 Pro · 24 GB
Runs well
est. 17.2 tok/s13.7–20.6calibrated estimate ±20%
Q4_K_M11.3 / 16.8 GBUnified memoryup to 16k——
Ryzen AI Max+ 395 (Strix Halo) · 32 GB
Runs well
est. 15.5 tok/s12.4–18.6calibrated estimate ±20%
Q4_K_M11.3 / 22.4 GBUnified memoryup to 16k—$1,999 launch MSRP
Apple M4 Pro · 24 GB
Runs well
est. 15.3 tok/s12.2–18.3calibrated estimate ±20%
Q4_K_M11.3 / 16.8 GBUnified memoryup to 16k——
GeForce RTX 3080 10GB
Runs slowly
est. 15.1 tok/s12.1–18.1calibrated estimate ±20%
Q4_K_M10.0 / 10.0 GB + 1.9 GB RAMPartial offloadup to 16k
  • Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly
$699 launch MSRP
NVIDIA DGX Spark · 128 GB
Runs slowly
est. 10.2 tok/s8.1–12.2calibrated estimate ±20%
Q4_K_M11.3 / 89.6 GBUnified memoryup to 16k—$3,999 launch MSRP
GeForce RTX 3070
Runs slowly
est. 8.1 tok/s6.5–9.8calibrated estimate ±20%
Q4_K_M8.0 / 8.0 GB + 3.9 GB RAMPartial offloadup to 16k—$499 launch MSRP
GeForce RTX 5060
Runs slowly
est. 8.1 tok/s6.5–9.8calibrated estimate ±20%
Q4_K_M8.0 / 8.0 GB + 3.9 GB RAMPartial offloadup to 16k—$339 street (as of 2026-08-09)
GeForce RTX 5060 Ti 8GB
Runs slowly
est. 8.1 tok/s6.5–9.8calibrated estimate ±20%
Q4_K_M8.0 / 8.0 GB + 3.9 GB RAMPartial offloadup to 16k—$429 street (as of 2026-08-09)
GeForce RTX 4060 Ti 8GB
Runs slowly
est. 7.5 tok/s6.0–9.0calibrated estimate ±20%
Q4_K_M8.0 / 8.0 GB + 3.9 GB RAMPartial offloadup to 16k—$399 launch MSRP
GeForce RTX 4060
Runs slowly
est. 7.4 tok/s5.9–8.9calibrated estimate ±20%
Q4_K_M8.0 / 8.0 GB + 3.9 GB RAMPartial offloadup to 16k—$299 launch MSRP
GeForce RTX 4060 Laptop
Runs slowly
est. 7.3 tok/s5.8–8.7calibrated estimate ±20%
Q4_K_M8.0 / 8.0 GB + 3.9 GB RAMPartial offloadup to 16k——
GeForce RTX 4070 Laptop
Runs slowly
est. 7.3 tok/s5.8–8.7calibrated estimate ±20%
Q4_K_M8.0 / 8.0 GB + 3.9 GB RAMPartial offloadup to 16k——
GeForce RTX 3050 8GB
Runs slowly
est. 7.0 tok/s5.6–8.4calibrated estimate ±20%
Q4_K_M8.0 / 8.0 GB + 3.9 GB RAMPartial offloadup to 16k—$249 launch MSRP
Apple M5 · 16 GB
Runs slowly
est. 5.0 tok/s3.5–6.5theoretical estimate ±30%
Q4_K_M10.7 GB RAMCPU onlyup to 8k——
DDR5-6000 dual-channel CPU · 16 GB
Runs slowly
est. 4.5 tok/s3.6–5.4calibrated estimate ±20%
Q4_K_M10.7 GB RAMCPU onlyup to 8k——
DDR5-5600 dual-channel CPU · 16 GB
Runs slowly
est. 4.2 tok/s3.3–5.0calibrated estimate ±20%
Q4_K_M10.7 GB RAMCPU onlyup to 8k——
Apple M4 · 16 GB
Runs slowly
est. 3.9 tok/s2.7–5.1theoretical estimate ±30%
Q4_K_M10.7 GB RAMCPU onlyup to 8k——
DDR4-3200 dual-channel CPU · 64 GB
Won't run
est. 2.4 tok/s1.9–2.9calibrated estimate ±20%
Q4_K_M10.7 GB RAMCPU only—
  • Fits, but too slow to use
—

Reasoning models spend extra tokens thinking, so their speed thresholds are 1.5× stricter (30 / 12 / 3 tok/s).

Cheapest GPUs that run it

  • Runs greatGeForce RTX 5070$629 street (as of 2026-08-09), est. 43.8 tok/s (38.6–49.1, calibrated estimate ±12%).
  • Runs wellGeForce RTX 3060 12GB$250 street (as of 2026-08-09), est. 23.5 tok/s (20.7–26.3, calibrated estimate ±12%).

Measured results for Phi-4

No public measurements for this model yet.

Frequently asked questions

How much VRAM does Phi-4 need?

At Q4_K_M the weights are 9.05 GB; with 8k context the total is about 11.3 GB (KV cache 1.7 GB, compute buffer 0.6 GB).

What is the cheapest GPU that runs Phi-4 well?

Cheapest that runs it great: GeForce RTX 5070 at $629 (street price as of 2026-08-09), est. 43.8 tok/s (38.6–49.1, calibrated estimate ±12%). Cheaper and still Runs well: GeForce RTX 3060 12GB at $250 (street price as of 2026-08-09), est. 23.5 tok/s (20.7–26.3, calibrated estimate ±12%).

Can Phi-4 run on an 8 GB, 16 GB or 24 GB GPU?

At Q4_K_M with 8k context: GeForce RTX 4060 — Runs slowly, est. 7.4 tok/s (5.9–8.9, calibrated estimate ±20%); GeForce RTX 5060 Ti 16GB — Runs well, est. 29.2 tok/s (25.7–32.7, calibrated estimate ±12%); GeForce RTX 3090 — Runs great, est. 61.1 tok/s (53.7–68.4, calibrated estimate ±12%).

Speeds are estimates from memory bandwidth, calibrated against public benchmarks, and each one comes with an error band and a confidence label. Real results vary with drivers, backend, context length and thermals.